Dynamic Dashboards Using Base SAS Software Kirk Paul Lafler, Software Intelligence Corporation, Spring Valley, California
Abstract Dynamic interactive visual displays known as dashboards are most effective when they show essential graphs, tables, statistics, and other information where data is the star. The first rule for creating an effective dashboard is to keep it simple. Striking a balance between content and style, a dashboard should be void of excessive clutter so as not to distract and obscure the information displayed. The second rule of effective dashboard design involves displaying data that meets one or more business or organizational objectives. To accomplish this, the elements in a dashboard should convey a format easily understood by its intended audience. Attendees learn how to create dynamic interactive user- and data-driven dashboards, graphical and table-driven dashboards, statistical dashboards, and drill-down dashboards with a purpose.
Introduction In a world of big data where data repositories and the demand placed on them are growing at explosive levels, organizations are faced with a number of decisions related to their information requirements: 1) What are the best ways to handle large amounts of information?
2) How should analytical data be processed?
3) What are the choices for constructing the most effective information delivery mechanisms?
4) How should analytical data and results be displayed?
To help answer these and other questions, this paper explains what a dashboard is, the dashboards elements, the dos and donts for constructing effective dashboards, dashboard design techniques, an investigation of the various types of dashboards, the merits and strengths of using the base-SAS software to construct dashboards, and an illustration of a few dashboard examples along with the base-SAS code used in their construction.
Example Table The dashboard examples displayed in this paper reference a Movies table consisting of a number of movies that Ive viewed over the years. The Movies table consists of six columns: Title, Length, Category, Year, Studio, and Rating. Title, Category, Studio, and Rating are defined as character columns, and the Length and Year are defined as numeric columns. The data contained in the Movies table is illustrated below.
Dynamic Dashboards Using Base SAS Software, continued SGF 2016
Brief History of Dashboards In the world of information technology, a dashboard serves as a user interface to organize and display information visually in the simplest way possible. Dashboards originated in the 1970s as decision support tools and systems that served management, operations, and organizational planning. In the 1980s, dashboards came of age as executive information systems emphasizing graphical displays and simple user interfaces to assist with management decision making. In the 1990s, dashboards experienced a growing interest with the rise of the Internet. As information technology and the Internet entered the 2000s, vendors including SAS Institute, and others, offered high-end easy-to-use products for the development of comprehensive custom dashboards. The dashboards being built today offer users the ability to monitor key metrics, information summaries, and reports in a single easy-to-use user interface. As a result, dashboards are designed to alert users to key business issues that impact an organizations tactics and strategies by facilitating improved decision making activities. So exactly what is a dashboard? In the paper, Building Your First Dashboard Using the SAS 9 Business Intelligence Platform: A Tutorial, by Gregory S. Nelson (2009), Nelson describes a dashboard as a visualization technique that provides an immediate view or snapshot of exactly where you are in a specific process relative to your stated goals and objectives. He adds that, Visual indicators, such as temperature gauges, traffic lights and speedometers, help give a real-world sense of present progress and assists in making decisions, adapting to current conditions or drilling into more detailed information. As a user interface, dashboards display performance indicators (PIs), key performance indicators (KPIs), and other relevant information.
Types of Dashboards The first step in dashboard design is to understand the purpose and type of dashboard you will need. With three types of dashboard designs available, users are encouraged to select the dashboard type that best meets your needs. The following table describes the three types of enterprise dashboards and their purpose.
Dashboard Type Purpose
Strategic Dashboards Strategic dashboards provide executives and managers with visual information to determine and support goals and objectives within an organization. This type of dashboard facilitates monitoring an organizations health, progress, performance, and areas where improvement can be made. There is typically no need for interactive features with this type of dashboard. Strategic dashboard examples include: Sales, Human Resources, Manufacturing, and Services.
Analytical Dashboards Analytical dashboards provide users with visual information to help gain a better understanding with historical, present and future data; understand trends; allow comparisons to be made; and determine the type of adjustments that are needed. Analytical dashboards should allow interactive features such as drill-down capabilities, as needed, to access more detailed information. Dashboard examples include: obtaining real-time data and information, determining why some things are working and others are not, identifying patterns and opportunities with your data, and aligning strategic objectives with performance initiatives.
Operational Dashboards Operational dashboards provide users with visual information to concentrate on performance monitoring and measurements, monitor the efficiency and effectiveness of their organization. There is typically a need to update information displayed in an operational dashboard frequently to make it relevant to the users needs. Dashboard examples include: improved understanding of performance, better focus and alignment, and faster and better decision making.
Dashboard Elements In Malik Shadans (2007) paper, Elements for an Enterprise Dashboard, he mentions that there are basic and advanced characteristics specific to an enterprise dashboard. The basic characteristics encompass the acronym, SMART, and the advanced characteristics of an enterprise dashboard encompass the acronym, IMPACT. The elements associated with each acronym appear in the following tables.
SMART Basic Elements Description
Synergetic Synergize information in a single screen view.
Monitor KPIs Display critical KPIs for effective decision making.
Accurate Dashboard must be well tested and validated, and information must be accurate.
Responsive Respond to user alerts and visual content to draw immediate attention to critical matters.
Timely Display information that is real-time and right-time for effective decision making.
Dynamic Dashboards Using Base SAS Software, continued SGF 2016
IMPACT Advanced Elements
Interactive Allow user to drill-down and derive details, root causes and more.
More Data History Allow users to review historical trends for any KPI.
Personalized Display should be specific to each users domain of responsibility, data restrictions, and privileges.
Analytical Allow users to perform guided analysis, compare, contrast, and make analytical inferences.
Collaborative Facilitate users ability to exchange notes regarding observations on their dashboard.
Trackability Allow each user to customize the metrics they would like to track.
Avoiding 13 Common Pitfalls in Dashboard Design Successful dashboard design involves the transformation of quantitative data into meaningful and effective visual displays including graphs, maps, gauges and summary information. In his paper, Common Pitfalls in Dashboard Design, Stephen Few (2006) proposes 13 common mistakes many make when designing dashboards. Instead of concentrating on what should be done when designing dashboards, Mr. Fews body of work espouses the most common mistakes along with detailed explanations to help educate current and future designers alike. I have listed the 13 common pitfalls from Mr. Fews seminal work, below, but readers are encouraged to read his entire paper, see the References section, for a complete perspective. Stephen Fews 13 Common Pitfalls in Dashboard Design (cited from reference)
Pitfall #1 Exceeding the Boundaries of a Single Screen
Pitfall #2 Supplying Inadequate Context for the Data
Pitfall #3 Displaying Excessive Detail or Precision
Pitfall #4 Expressing Measures Indirectly
Pitfall #5 Choosing Inappropriate Media of Display
Pitfall #6 Introducing Meaningless Variety
Pitfall #7 Using Poorly Designed Display Media
Pitfall #8 Encoding Quantitative Data Inaccurately
Pitfall #9 Arranging the Data Poorly
Pitfall #10 Ineffectively Highlighting Whats Important
Pitfall #11 Cluttering the Screen with Useless Decoration
Pitfall #12 Misusing or Overusing Color
Pitfall #13 Designing an Unappealing Visual Display
Steps to Creating a Dynamic Dashboard using Base-SAS Follow these basic steps to successfully construct a dynamic dashboard using the Base-SAS software.
1. Connect to desired data sources using Libname statement.
2. Create user-defined formats containing URL links for dashboard and drill-down results.
3. Create Graphics, Bar Chart, Box Plot, Histogram, Pie, etc.